Updated: tesseract-ocr-3.04.00-2

[Marco Atzeri]

Version 3.04.00-2 of packages

libtesseract-ocr_3
tesseract-ocr
tesseract-ocr-devel
tesseract-training-util (NEW)

and version 3.04-1 of

tesseract-ocr-eng
tesseract-ocr-deu
tesseract-ocr-fra
tesseract-ocr-ita
tesseract-ocr-nld
tesseract-ocr-por
tesseract-ocr-spa
tesseract-ocr-vie

tesseract-training-core (NEW)
tesseract-training-eng (NEW)
tesseract-training-deu (NEW)
tesseract-training-fra (NEW)
tesseract-training-ita (NEW)
tesseract-training-nld (NEW)
tesseract-training-por (NEW)
tesseract-training-spa (NEW)
tesseract-training-vie (NEW)

are available in the Cygwin distribution:

Other language specific data are available upstream
https://github.com/tesseract-ocr/tessdata

while training data for building new language data are in
https://github.com/tesseract-ocr/langdata


CYGWIN CHANGES
Rebuilt to include the training tools and base data
to create or update language data.

Training tools, not needed for normal users, are in
tesseract-training-util

and data in
tesseract-training-core
tesseract-training-{lang}


CHANGES
None. Last upstream release.
https://github.com/tesseract-ocr/tesseract/wiki


DESCRIPTION
Tesseract is probably the most accurate open source OCR engine
available. Combined with the Leptonica Image Processing Library
it can read a wide variety of image formats and convert them to
text in over 60 languages. It was one of the top 3 engines in
the 1995 UNLV Accuracy test.
Improved extensively by Google.
It is released under the Apache License 2.0.


HOMEPAGE
https://github.com/tesseract-ocr/


Marco Atzeri

If you have questions or comments, please send them to the
cygwin mailing list at: cygwin (at) cygwin (dot) com .